Ontology Alignment state of the art and an application in literature search Patrick Lambrix Linköpings universitet.

Ontology Alignmentstate of the art andan application in literature search

Patrick Lambrix

Linköpings universitet

Ontologies

“Ontologies define the basic terms and relations comprising the vocabulary of a topic area, as well as the rules for combining terms and relations to define extensions to the vocabulary.”

(Neches, Fikes, Finin, Gruber, Senator, Swartout, 1991)

Example GENE ONTOLOGY (GO)

immune response i- acute-phase response i- anaphylaxis i- antigen presentation i- antigen processing i- cellular defense response i- cytokine metabolism i- cytokine biosynthesis synonym cytokine production … p- regulation of cytokine biosynthesis … … i- B-cell activation i- B-cell differentiation i- B-cell proliferation i- cellular defense response … i- T-cell activation i- activation of natural killer cell activity …

Ontologies used …

for communication between people and organizations

for enabling knowledge reuse and sharing as basis for interoperability between systems as repository of information as query model for information sources

Key technology for the Semantic Web

Biomedical Ontologies - efforts

OBO – Open Biomedical Ontologieshttp://www.obofoundry.org/(over 50 ontologies)

” The mission of OBO is to support community members who are developing and publishing ontologies in the biomedical domain. It is our vision that a core of these ontologies will be fully interoperable, by virtue of a common design philosophy and implementation, thereby enabling scientists and their instruments to communicate with minimum ambiguity. In this way the data generated in the course of biomedical research will form a single, consistent, cumulatively expanding, and algorithmically tractable whole. This core will be known as the "OBO Foundry". .”

OBO Foundry

1. open and available 2. common shared syntax 3. unique identifier space 4. procedures for identifying distinct successive versions5. clearly specified and clearly delineated content6. textual definitions for all terms7. use relations from OBO Relation Ontology8. well documented 9. plurality of independent users10. developed collaboratively with other OBO Foundry

members

Biomedical Ontologies - efforts

National Center for Biomedical Ontology http://bioontology.org/index.html

Funded by National Institutes of Health

”The goal of the Center is to support biomedical researchers in their knowledge-intensive work, by providing online tools and a Web portal enabling them to access, review, and integrate disparate ontological resources in all aspects of biomedical investigation and clinical practice. A major focus of our work involves the use of biomedical ontologies to aid in the management and analysis of data derived from complex experiments.”

Systems Biology Ontologies - efforts

Systems Biology Ontology Proteomics Standard Initiative for Molecular

Interaction BioPAX

Ontology Alignment

Ontology alignmentOntology alignment Ontology alignment strategies Evaluation of ontology alignment strategies Current issues Ontology-based literature search

Ontologies in biomedical research

many biomedical ontologies

practical use of biomedical ontologiese.g. databases annotated with GO

GENE ONTOLOGY (GO)


Ontologies with overlapping information

SIGNAL-ONTOLOGY (SigO)

Immune Response i- Allergic Response i- Antigen Processing and Presentation i- B Cell Activation i- B Cell Development i- Complement Signaling synonym complement activation i- Cytokine Response i- Immune Suppression i- Inflammation i- Intestinal Immunity i- Leukotriene Response i- Leukotriene Metabolism i- Natural Killer Cell Response i- T Cell Activation i- T Cell Development i- T Cell Selection in Thymus

GENE ONTOLOGY (GO)


Ontologies with overlapping information Use of multiple ontologies

e.g. custom-specific ontology + standard ontology different views on same domain connecting related areas

Bottom-up creation of ontologiesexperts can focus on their domain of expertise

important to know the inter-ontology important to know the inter-ontology relationshipsrelationships



GENE ONTOLOGY (GO)


Ontology Alignment

equivalent concepts

equivalent relations

is-a relation



GENE ONTOLOGY (GO)


Defining the relations between the terms in different ontologies

Ontology Alignment

Ontology alignment Ontology alignment strategiesOntology alignment strategies Evaluation of ontology alignment strategies Current issues Ontology-based literature search

An Alignment Framework

Preprocessing

Preprocessing

For example, Selection of features Selection of search space

Matchers

Strategies based on linguistic matching Structure-based strategies Constraint-based approaches Instance-based strategies Use of auxiliary information

Matcher Strategies

Strategies based on linguistic matchingStrategies based on linguistic matching

SigO: complement signaling synonym complement activation

GO: Complement Activation

Example matchers

Edit distance Number of deletions, insertions, substitutions required to transform one

string into another aaaa baab: edit distance 2

N-gram N-gram : N consecutive characters in a string Similarity based on set comparison of n-grams aaaa : {aa, aa, aa}; baab : {ba, aa, ab}

Matcher Strategies

Strategies based on linguistic matching Structure-based strategiesStructure-based strategies Constraint-based approaches Instance-based strategies Use of auxiliary information

Example matchers

Propagation of similarity values Anchored matching

Example matchers


Example matchers


Matcher Strategies

Strategies based on linguistic matching Structure-based strategies Constraint-based approachesConstraint-based approaches Instance-based strategies Use of auxiliary information

O1O2

Bird

Mammal Mammal

FlyingAnimal

Matcher Strategies

Strategies based on linguistic matching Structure-based strategies Constraint-based approachesConstraint-based approaches Instance-based strategies Use of auxiliary information

O1O2

Bird

Mammal Mammal

Stone

Example matchers

Similarities between data types Similarities based on cardinalities

Matcher Strategies

Strategies based on linguistic matching Structure-based strategies Constraint-based approaches Instance-based strategiesInstance-based strategies Use of auxiliary information

Ontology

instancecorpus

Example matchers

Instance-based Use life science literature as instances

Learning matchers – instance-based strategies

Basic intuition A similarity measure between concepts can be

computed based on the probability that documents about one concept are also about the other concept and vice versa.

Basic Naïve Bayes matcher

Generate corpora Use concept as query term in PubMed Retrieve most recent PubMed abstracts

Generate classifiers Naive Bayes classifiers, one per ontology

Classification Abstracts related to one ontology are classified to the concept

in the other ontology with highest posterior probability P(C|d)

Calculate similarities

Matcher Strategies

Strategies based linguistic matching Structure-based strategies Constraint-based approaches Instance-based strategies Use of auxiliary informationUse of auxiliary information

thesauri

alignment strategies

dictionary

intermediateontology

Example matchers

Use of WordNet Use WordNet to find synonyms Use WordNet to find ancestors and descendants in the is-

a hierarchy

Use of Unified Medical Language System (UMLS) Includes many ontologies Includes many mappings (not complete) Use UMLS mappings in the computation of the

similarity values

Ontology A

lignment and M

ergning S

ystems

Combinations

Combination Strategies

Usually weighted sum of similarity values of different matchers

Maximum of similarity values of different matchers

Filtering

Threshold filtering Pairs of concepts with similarity higher or equal t

han threshold are mapping suggestions

Filtering techniques

th

( 2, B )

( 3, F )

( 6, D )

( 4, C )

( 5, C )

( 5, E )

……

suggest

discard

sim

Filtering techniques

lower-th

( 2, B )

( 3, F )

( 6, D )

( 4, C )

( 5, C )

( 5, E )

……

upper-th

Double threshold filtering(1) Pairs of concepts with similarity higher than or equal to upper threshold are

mapping suggestions

(2) Pairs of concepts with similarity between lower and upper thresholds are mapping suggestions if they make sense with respect to the structure of the ontologies and the suggestions according to (1)

Example alignment system SAMBO – preprocessing, matchers, combination, filter

Example alignment system SAMBO – suggestion mode

Example alignment system SAMBO – manual mode

Ontology Alignment

Ontology alignment Ontology alignment strategies Evaluation of ontology alignment strategies Evaluation of ontology alignment strategies Current issues Ontology-based literature search

Evaluation measures Precision: # correct suggested mappings # suggested mappings Recall: # correct suggested mappings # correct mappings F-measure: combination of precision and

recall

Ontology AlignmentEvaluation Initiative

OAEI

Since 2004 Evaluation of systems Different tracks

comparison: benchmark (open) expressive: anatomy (blind), fisheries (expert) directories and thesauri: directory, library, cros

slingual resources (blind) consensus: conference

OAEI 2007 17 systems participated

benchmark (13) ASMOV: p = 0.95, r = 0.90

anatomy (11) AOAS: f = 0.86, r+ = 0.50 SAMBO: f =0.81, r+ = 0.58

library (3) Thesaurus merging: FALCON: p = 0.97, r = 0.87 Annotation scenario:

FALCON: pb =0.65, rb = 0.49, pa = 0.52, ra = 0.36, Ja = 0.30 Silas: pb = 0.66, rb= 0.47, pa = 0.53, ra = 0.35, Ja = 0.29

directory (9), food (6), environment (2), conference (6)

OAEI 2008 – anatomy track Align

Mouse anatomy: 2744 terms NCI-anatomy: 3304 terms Mappings: 1544 (of which 934 ‘trivial’)

Tasks 1. Align and optimize f 2-3. Align and optimize p / r 4. Align when partial reference alignment is

given and optimize f

OAEI 2008 – anatomy track#1

9 systems participated SAMBO

p=0.869, r=0.836, r+=0.586, f=0.852 SAMBOdtf

p=0.831, r=0.833, r+=0.579, f=0.832 Use of TermWN and UMLS

OAEI 2008 – anatomy track#1

Is background knowledge (BK) needed?

Of the non-trivial mappings: Ca 50% found by systems using BK and systems not

using BK Ca 13% found only by systems using BK Ca 13% found only by systems not using BK Ca 25% not found

Processing time: hours with BK, minutes without BK

OAEI 2008 – anatomy track#4Can we use given mappings when computing suggestions? partial reference alignment given with all trivial and 50 n

on-trivial mappings

SAMBO p=0.6360.660, r=0.6260.624, f=0.6310.642

SAMBOdtf p=0.5630.603, r=0.6220.630, f=0.5910.616

(measures computed on non-given part of the reference alignment)

OAEI 2007-2008

Systems can use only one combination of strategies per task

systems use similar strategies text: string matching, tf-idf structure: propagation of similarity to ancestors

and/or descendants thesaurus (WordNet) domain knowledge important for anatomy task?

Ontology Alignment

Ontology alignment Ontology alignment strategies Evaluation of ontology alignment strategies Current IssuesCurrent Issues Ontology-based literature search

Current issues

Systems and algorithms Complex ontologies Use of instance-based techniques Alignment types (equivalence, is-a, …) Complex mappings (1-n, m-n) Connection ontology types – alignment strategies

EvaluationSEALS – Semantic Evaluation At Large Scale

Current issues

Recommending ’best’ alignment strategies

Use of Partial Reference Alignment

--------------------------------------------------------- Integration of ontology alignment and repair of

the structure of ontologies

Ontology Alignment

Ontology alignment Ontology alignment strategies Evaluation of ontology alignment strategies Current issues Ontology-based literature searchOntology-based literature search

Literature search

Huge amount of scientific literature.

Need to integrate a spectrum of information to perform a task.

Literature search

How to know what is in the repositoryLack of knowledge of the domain

How to compose an expressive queryLack of knowledge of search technology

Example scenario“Lipid” Keyword search returns all documents

containing lipid.No knowledge; terminology problem

Relationships: use of multiple keywords with/without boolean operators, e.g. lipid and disease

Example scenario“Lipid” Keyword search returns a list of relevant

questions concerning lipid. User selects question and retrieves knowledge and provenance documents.

Multiple search terms: requirement that there are relevant connections between the keywords.

lipid

Relevant queries

Relevant query including a number of concepts and relations from an ontology

connected sub-graph of the ontology that includes the concepts and relations.

(query graph based on the concepts and relations;

slice is set of all query graphs based on the concepts and relations)

Query graph

Query graph

Query graph

Special cases

No relations, several concepts Relevant queries regarding concepts; relations are

suggested by the system. Difference with traditional techniques: extra requirement

that search terms need to be connected in the ontology.

No relations, one concept Relevant queries including a specific query term. Computes the ontological environment of the query term.

Relevant queries – multiple ontologies

Relevant query including a number of concepts and relations from multiple ontologies

Query graphs connected by a path going through a mapping in the alignment.

(aligned query graph based on query graphs; aligned slice is set of all aligned query graphs

based on the query graphs)

Aligned query graph

Aligned query graph

Aligned query graph

Framework

External resources

Literature document baseGenerated from a collection of 7498 PubMed

abstracts relevant for Ovarian Cancer. 683 papers included lipid names from which 241 full papers were downloadable.

Ontology and ontology alignment repositoryLipid ontologySignal ontologyAligment using SAMBO

2) Sentence Extraction

1) Document Content

3) Sentence Detection: lipid interaction protein

4) Entity Recognition: term identification / assign lipid class

5) Normalization: collapse lipid synonyms

6) Relation Extraction: Lipid-Protein or Lipid Disease

8) Populate OWL ontology (JENA -API)

Complete Instantiated

OWL-DLOntology

Term List DB’s: Lipid names, LIPIDMAPS, Lipid Bank, KEGG classifications, Disease names, Protein names Stemmed Interactions

Document and sentence meta data "TLR4 binds to POPC", tagged as "TLR4 binds to POPC", tagged as

"<term category=""<term category="proteinprotein"> TLR4</term> "> TLR4</term> binds to binds to <term category="<term category="lipidlipid">POPC</term>"">POPC</term>"

7) Classification: Identify ontology classes and specify relations for all sentences, proteins, lipid subclasses.

Knowledge base instantiation

Lipid Instance

Lipid Instance

Lipid Class Protein Instance

Knowledge base instantiation

Slice generation

Current implementation focuses on slices based on concepts.

Depth-first traversal of ontology to find paths between given concepts; paths can be put together to find slices/query graphs.

Slice alignment

Algorithm computes subset of aligned slice. Assumption: shorter paths represent closer

relationships. Algorithm connects slices using shortest paths

from given concepts in one ontology to given concepts in other ontology.

Slicing through the literature

protein lipid diseaseSignal-pathway

Involved-in Interacts-with Implicated-in

Natural language query generation

Triple representation:

<lipid, interacts-with, protein>

Rule base to generate NL statements.

What lipid interacts with proteins?Learned from examples.

Aggregation of statements from different triples, grammar checking.

Query

Send nRQL query to RACER.

Future Work

Tradeoff in query generation between completeness and information overload.

Relevance measure and query ranking. Integrated implementation. Scalability testing.

Further reading

Ontology alignment - general

http://www.ontologymatching.org(plenty of references to articles and systems)

Ontology alignment evaluation initiative: http://oaei.ontologymatching.org(home page of the initiative)

Euzenat, Shvaiko, Ontology Matching, Springer, 2007.

Lambrix, Strömbäck, Tan, Information integration in bioinformatics with ontologies and standards, in Bry, Maluszynski (eds), Semantic Techniques for the Web: The REWERSE perspective, chapter 8, 343-376, 2009.

(contains currently largest overview of ontology alignment systems)

Further reading

Ontology alignment - systems Lambrix, Tan, SAMBO – a system for aligning and merging biomedical

ontologies, Journal of Web Semantics, 4(3):196-206, 2006.(description of the SAMBO tool and overview of evaluations of different

matchers)

Lambrix, Tan, A tool for evaluating ontology alignment strategies, Journal on Data Semantics, VIII:182-202, 2007.

(description of the KitAMO tool for evaluating matchers)

Further readingOntology alignment - recommendation of alignment strategies Tan, Lambrix, A method for recommending ontology alignment strategies,

International Semantic Web Conference, 494-507, 2007. Ehrig, Staab, Sure, Bootstrapping ontology alignment methods with

APFEL, International Semantic Web Conference, 186-200, 2005. Mochol, Jentzsch, Euzenat, Applying an analytic method for matching

approach selection, International Workshop on Ontology Matching, 2006.

Ontology alignment - PRA in ontology alignment Lambrix, Liu, Using partial reference alignments to align ontologies,

European Semantic Web Conference, 188-202, 2009.

Literature search Baker, Lambrix, Laurila Bergman, Kanagasabai, Ang, Slicing through the

scientific literature, Data Integration in the Life Sciences, 127-140, 2009.

DILS 20107th International Conference on Data Integration in the Life Sciences

August 25-27, Gothenburg, Sweden

paper submission deadline in April

Ontology Alignment state of the art and an application in literature search Patrick Lambrix Linköpings universitet.

Documents

ontologies ontologies

use of biomedical ontologies

biomedical ontology

publishing ontologies

biomedical researchers

biomedical domain

cell activation i activation

cell activation i b